SESSION C-4 : 교통 데이터 분석 및 AI
※ 기관로그인 시 무료 이용이 가능합니다.
Abstract.
I. Introduction
1. Survey Inconsistency
2. Heuristic Cleaning and Information Loss
3. The Recursive Neurosymbolic Optimizer(RNO)
II. Literature Review
1. Architectural Roles of LLMs in Mobility Modelling
2. Quantitative Performance Benchmarks
3. Physical Validity and Symbolic Constraints
4. Mitigation of Data Sparsity and Noise
5. Methodological Challenges and the NeurosymbolicGap
III. Methodology
1. Overview: The Recursive Neurosymbolic Optimizer (RNO)
2. Symbolic Layer: The Auditor
3. Neural Layer: The Generator
4. Recursive Feedback and Temporal Elasticity
5. Batch Processing and Post-Processing
6. Validation Protocol
IV. Results
1. Structural Health Score Progression
2. Trip-Count Convergence
3. Distributional Fidelity: Jensen-Shannon Divergence
4. Batch Performance
5. Final Dataset
6. Iterative Convergence and Temporal Elasticity
V. Discussion
1. Framework Portability and Architectural Consistency
2. Information Preservation vs. Heuristic Deletion
3. Population Heterogeneity and DistributionalDivergence
4. Urban Planning Impact: Axiom-Compliant Dataset for Simulation
5. Recursive Convergence and Global Rescheduling Authority
6. Limitations and Future Work
VI. Conclusion
VII. References
0개의 논문이 장바구니에 담겼습니다.